In deduplication storage systems, duplicate data is identified so that ideally only a single copy of the data is actually stored. Later copies of the duplicate data merely reference or otherwise point to the already-stored data so that duplicates of the same piece of data are not stored. Although deduplication storage systems exist, new deduplication storage systems which are easy to expand, but do not significantly compromise on the amount of storage saved by deduplication, would be desirable.